app.py

import os
import gradio as gr
from dotenv import load_dotenv
from openai import OpenAI
from prompts.initial_prompt import INITIAL_PROMPT
from prompts.main_prompt import (
    MAIN_PROMPT,
    get_prompt_for_problem,
    get_ccss_practice_standards,
    get_problem_posing_task,
    get_creativity_discussion,
    get_summary,
)

# Load API key from .env file
if os.path.exists(".env"):
    load_dotenv(".env")

OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
client = OpenAI(api_key=OPENAI_API_KEY)

def gpt_call(history, user_message, model="gpt-4o-mini", max_tokens=512, temperature=0.7, top_p=0.95):
    """
    Calls OpenAI Chat API to generate responses.
    - history: [(user_text, assistant_text), ...]
    - user_message: latest message from user
    """
    messages = [{"role": "system", "content": MAIN_PROMPT}]

    # Add history to conversation
    for user_text, assistant_text in history:
        if user_text:
            messages.append({"role": "user", "content": user_text})
        if assistant_text:
            messages.append({"role": "assistant", "content": assistant_text})

    messages.append({"role": "user", "content": user_message})

    completion = client.chat.completions.create(
        model=model,
        messages=messages,
        max_tokens=max_tokens,
        temperature=temperature,
        top_p=top_p
    )

    return completion.choices[0].message.content

def respond(user_message, history):
    """
    Handles user input and chatbot responses.
    - user_message: latest user input
    - history: previous chat history
    """
    if not user_message:
        return "", history

    # If user selects a problem number, redirect to the appropriate prompt
    if user_message.strip() in ["1", "2", "3"]:
        assistant_reply = get_prompt_for_problem(user_message.strip())

    # If user is at reflection stage, ask about CCSS Practice Standards
    elif user_message.lower().strip() == "common core":
        assistant_reply = get_ccss_practice_standards()

    # If user is at problem-posing stage, ask them to create a new problem
    elif user_message.lower().strip() == "problem posing":
        assistant_reply = get_problem_posing_task()

    # If user is at creativity discussion stage, ask for their thoughts
    elif user_message.lower().strip() == "creativity":
        assistant_reply = get_creativity_discussion()

    # If user requests a summary, provide the final learning summary
    elif user_message.lower().strip() == "summary":
        assistant_reply = get_summary()

    else:
        # Continue conversation normally with AI guidance
        assistant_reply = gpt_call(history, user_message)

    # Update history
    history.append((user_message, assistant_reply))
    return "", history

##############################
#  Gradio UI Setup
##############################
with gr.Blocks() as demo:
    gr.Markdown("## AI-Guided Math PD Chatbot")

    # Initialize chatbot with first message
    chatbot = gr.Chatbot(
        value=[("", INITIAL_PROMPT)],  # Initial system message
        height=500
    )

    # Maintain chat history state
    state_history = gr.State([("", INITIAL_PROMPT)])

    # User input box
    user_input = gr.Textbox(
        placeholder="Type your message here...",
        label="Your Input"
    )

    # Submit button
    user_input.submit(
        respond,
        inputs=[user_input, state_history],
        outputs=[user_input, chatbot]
    ).then(
        fn=lambda _, h: h,
        inputs=[user_input, chatbot],
        outputs=[state_history]
    )

# Launch app
if __name__ == "__main__":
    demo.launch(server_name="0.0.0.0", server_port=7860, share=True)